CN1742249A - Power management system - Google Patents
Power management system Download PDFInfo
- Publication number
- CN1742249A CN1742249A CN200380107857.3A CN200380107857A CN1742249A CN 1742249 A CN1742249 A CN 1742249A CN 200380107857 A CN200380107857 A CN 200380107857A CN 1742249 A CN1742249 A CN 1742249A
- Authority
- CN
- China
- Prior art keywords
- assembly
- computation module
- power
- power consumption
- state
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 claims description 35
- 238000007726 management method Methods 0.000 claims description 29
- 230000008569 process Effects 0.000 claims description 23
- 230000005059 dormancy Effects 0.000 claims description 18
- 238000011084 recovery Methods 0.000 claims description 10
- 238000004519 manufacturing process Methods 0.000 claims description 8
- 238000004458 analytical method Methods 0.000 claims description 6
- 230000004048 modification Effects 0.000 claims description 6
- 238000012986 modification Methods 0.000 claims description 6
- 238000013500 data storage Methods 0.000 claims description 3
- 230000004044 response Effects 0.000 claims 1
- 238000004891 communication Methods 0.000 description 9
- 238000010586 diagram Methods 0.000 description 5
- 230000003716 rejuvenation Effects 0.000 description 5
- 230000008901 benefit Effects 0.000 description 3
- 239000004744 fabric Substances 0.000 description 3
- 238000012423 maintenance Methods 0.000 description 3
- 238000005192 partition Methods 0.000 description 3
- 230000000712 assembly Effects 0.000 description 2
- 238000000429 assembly Methods 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 2
- 230000004913 activation Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000000717 retained effect Effects 0.000 description 1
- 230000009897 systematic effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F1/00—Details not covered by groups G06F3/00 - G06F13/00 and G06F21/00
- G06F1/26—Power supply means, e.g. regulation thereof
- G06F1/32—Means for saving power
- G06F1/3203—Power management, i.e. event-based initiation of a power-saving mode
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Abstract
A power management system for a multiple compute component system. The management system enables operational management of individual compute components in two low power states of operation, suspend and hibernate. In addition, the management system enables granular management of power consuming accessories within an individual compute component. When a compute component is in the low power state of suspend or hibernate, it may be hot-swapped to a different locale. In addition, when a compute component is restored from a low power state of hibernate, it may access an alternate image in the system. Accordingly, the power management system allows for five state of operation of an individual component in a multiple compute component system.
Description
Technical field
This invention relates generally to the apparatus and method that managing electrical power consumes in computing environment.Or rather, the present invention relates to each computation module of management in many computation modules environment.
Background technology
Laptop computer is a personal computer, generally comprises microprocessor, hardware store, I/O port and communication port.Described laptop computer can move independently to satisfy the individual and calculate needs, if described laptop computer comprises network adapter or modulator-demodular unit, it also can become the part of local or wide area network.
One of current above-knee and characteristic that personal computer shows is that described computing machine can be placed in lower powered running status, such as hanging up and dormancy.These two kinds of low power operating states all are that personal computer is relevant known in the industry.Under described suspended state, to other annex powered-downs on processor and the mainboard, but internal memory remains unchanged.This is a kind of low power consumpting state.When the operator of described computing machine wishes to recover to use described hardware accessories, described operator must recover total power to described processor, described mainboard and described associated hardware accessories, although, needn't restart total system because described internal memory remains unchanged.Except described suspended state, described personal computer can also be placed in the dormant state of low-power consumption, being placed in the described internal memory after the disk, closes the power supply of described computing machine.So, enter described suspended state and the power supply that recovers described computing machine, perhaps enter the power supply of described dormant state and the described computing machine of recovery, rather than stop described computer power supply and restarted described total system afterwards, consuming time in general less.So using described suspended state on described above-knee or personal computer is an example of power management on the personal computer system.
According to definition, multiprocessor computer system comprises a plurality of processors, and this paper is also referred to as CPU, and they can carry out a plurality of threads in a plurality of processes or the single process simultaneously in the mode that is called parallel computation.In general, multicomputer system is carried out a plurality of processes or thread and is carried out the conventional single processor system of some programs faster than order, such as personal computer (PC).The actual performance advantage depends on many factors, comprises the degree that the several portions of multithreading process and/or a plurality of various process can executed in parallel, and the framework of concrete multicomputer system on hand.
How the framework of shared memory multiprocessor system can physically be organized according to its internal memory is classified.In distributed shared memory (DSM) machine, described internal memory is divided into plurality of modules, physically places near one or more processors, is in typical case on processor node.Although all described DRAM dies all are that the overall situation is addressable, this machine internal memory on its node of processor access can be faster than the long-distance inner on other nodes.Because the described internal storage access time, this type systematic was also referred to as non-homogeneous internal storage access (NUMA) machine according to the core position and difference.Otherwise in centralized shared drive machine, described internal memory physically is in a position.Centralized shared drive computing machine is called as even internal storage access (UMA) machine, because described internal memory is that the time is equidistant to each described processor.In typical case, two kinds of memory organization forms all use high-speed cache to assist main memory, to reduce the execution time.
Processor node can divide into groups to form subregion, and it is the set of the operating system next one or a plurality of node interconnections, to form computing environment.In same computer system, can there be a plurality of subregions.Each subregion in the computer system is carried out single SOS reflection.Multiprocessor computer system can be in the structure of a plurality of nodes or subregion set, comprises service processor hardware, supervisor console and other infrastructure, represents single managing and configurable environment.So a system also can be split into a plurality of logical machine systems or subregion, each all carries out single operation system image in them.
Except the multiple processor computation system of subregion joint form, also has the blade type multiple processor computation system.Described blade system is the set of Distributed Calculation resource available on LAN (Local Area Network) or the wide area network, to the final user with use and be rendered as a big virtual computing system.Every computational resource all is the server on the detachable card, is inserted in to share in the infrastructure, inserts frame again.Described computational resource can be shared keyboard, monitor and to the connection of described LAN (Local Area Network) or Wide Area Network.Described intrasystem every resource can be configured to move under different operating system.So the blade type multiple processor computation system is the scalable system example that has multinomial resource, described resource is suitable for communicating communication by public communicating to connect.
Partition type multiprocessor computing environment and blade type multiprocessor computing environment all comprise a plurality of computation modules.Each assembly comprises a slice printed circuit board (PCB) at least, has one or more microprocessors and a communication port.The current management that many computation modules system---comprises blade type computer system and partition type computer system---need be closed the computation module of appointment when needs are safeguarded.Each assembly all operates in and opens or closes a kind of in this two states.The intermediateness that does not have operation.This power management that has influence on assembly is the power consumption annex on the assembly in other words, and the plan of assembly and unplanned maintenance.So the intermediateness that is necessary to enable assembly operating in the computer system is so that can carry out effective power management of assembly in network computing environment.
Summary of the invention
The present invention includes the power-supply management system in many computation modules computing environment.For extensive and small-scale operation rank, described system all carries out power management.
In a first aspect of the present invention, provide the method for managing power supply in a kind of computation module.Described method comprise described assembly power consumption analysis, judge the power consumption level of described assembly needs and the running status of revising described assembly accordingly according to described power consumption analysis.Described method also can comprise to be controlled the granular management of power consumption annex in the described assembly.This class annex can comprise microprocessor, disc driver, I/O port and/or chip.The step of revising described assembly operating state can comprise by switch or software management system.The running status of described computation module comprise peak power, hardware accessories selectivity minimizing, hang-up, dormancy, recovery, recover and close.When described computation module when low power operating state returns, the alternate image that described computation module also can the accesses network memory device.In addition, when described assembly just when low power state moves, described assembly can carry out heat interchange.Lower powered running status comprises suspended state and dormant state.
In a second aspect of the present invention, provide multiple processor computation system with a plurality of computation modules.Manager is analyzed the power consumption of described assembly and is suitable for responding described analysis, calls the modification of described assembly operating state.The modification of running status can comprise the granular management of the power consumption annex of described assembly, such as microprocessor, disc driver, I/O port and chip.Described running status can comprise peak power, power selectivity minimizing, hang-up, dormancy, recovery, recover and close.In addition, described system can also comprise storage manager, is suitable for indicating described assembly during recovering described component function, the alternate image that visit is stored in the network storage equipment.Described assembly just when low power state moves, during such as described hang-up and dormant state, also can carry out heat exchanging process.
In a third aspect of the present invention, provide and contained manufacturing a product of computer-readable signal bearing medium.Described manufacturing a product comprises manager, is used for analyzing the power consumption of many computation modules computing environment computation module.In described medium, provide control system.Described control system responds described manager, changes the running status of described computation module.The form of described medium can be recordable data storage medium, or the carrier signal after the modulation.The running status of described assembly can comprise peak power, hardware accessories selectivity minimizing, hang-up, dormancy, recovery, recover or close.Described control system can be suitable for controlling the operation of power consumption annex in the described computation module.Under the preferable case, described power consumption annex comprises microprocessor, disc driver, I/O port and/or chip.In addition, described assembly is when dormant state is returned, and described control system can be connected to alternate image in the storer on the network for described computation module visit.When described assembly just when low power state moves, described control system also is suitable for providing the heat interchange of described computation module.
Together with the following detailed description that accompanying drawing is read the present invention's present preferred embodiment, other characteristics of the present invention and advantage will become apparent.
Brief Description Of Drawings
But Fig. 1 is the block diagram of prior art subregion multiprocessor computer system;
Fig. 2 is the synoptic diagram of prior art blade type multiprocessor computer system;
Fig. 3 is a width of cloth process flow diagram, has showed the granularity power management according to preferred embodiment of the present invention computation module, and suggestion is printed on first page of the patent of issuing;
Fig. 4 is a width of cloth process flow diagram, has showed the selection course of computation module low power operating state.
Fig. 5 is the synoptic diagram of multicompartment computer system, has showed a kind of substitute electric power management tool
Embodiment
Summary
The power management of each computation module has all strengthened operational efficiency to described each assembly and described system in many computation modules system.The supervisory level of each assembly can be that whole assembly is placed low power operating state.In addition, the mode of management of each assembly also can be the granular management of power consumption annex in the described computation module.The power management of two kinds of forms has all strengthened the efficient and the operation of described assembly and described system on the whole.
Ins and outs
Shown in Figure 1, multicomputer system 10 can comprise a plurality of nodes.Described system 10 uses distributed shared memory (DSM) framework, and it can be also can not be non-homogeneous internal storage access (NUMA) machine.As shown in Figure 1,12,14,16 and 18 4 nodes are arranged, each node is all connected by system interconnection 20, and it allows any node and any other node communication of described system.The purpose of described system interconnection 20 is to make the processor of any node can both visit resident internal memory on any other node of described system.The physical link of system interconnection 20 provides high bandwidth and low the delay, and is upgradeable to allow the more node of increase in described system 10.So, but described multicomputer system 10 is examples of zoned computer system.
But but the form of the described computation module in the described system is subregion, blade or equivalence subregion computation module in the partition system.Described assembly is a printed circuit board (PCB), comprises one or more microprocessors and a communication port, and it can comprise hardware accessories alternatively, such as storage hardware, internal memory, I/O port and battery.Fig. 2 is the example of multi-component system 30, and assembly 40 and 50 is stored in the frame 32.Assembly 140 is printed circuit board (PCB)s 42, is shown as to have hard disk 42 and charged pool not.Assembly 250 is printed circuit board (PCB)s 52, is shown as to have battery 54, but is not with hard disk.Assembly 140 and assembly 250 all have communication port (not shown) separately.Described assembly 40 and 50 shares to the order wire 34 of LAN (Local Area Network) 36.As shown in the figure, described LAN (Local Area Network) 36 all communicates with assembly 140 and assembly 250.Described frame 32 is shells, is suitable for storing an assembly or a plurality of assembly.The described order wire 34 that all described assemblies in described frame 32 are shared keyboard 60, display device 65 and arrived described LAN (Local Area Network) 36.Described frame 32 also is used for providing external power source to each described assembly.Assembly 40 and 50 all is shown as by described LAN (Local Area Network) 36 and is connected to the storer 55 that is connected on the network.Because assembly 140 has the hard disk of itself, it can store data in this locality in the storage hardware of itself, and the described storer 55 that is connected on the network also can write direct its canned data.In addition, assembly 140 also can use its visit to described LAN (Local Area Network) 36, with data storage to the center.
In the prior art, each computation module of multicompartment computer system can both operate in one of peak power or two kinds of running statuses of inactivity.In described preferred embodiment, each assembly can operate in one of five kinds of states, and two kinds of described states are peak power or inactivity, the intermediateness of three kinds of remaining states for moving.First kind of described three kinds of intermediate state of operation is the selectivity minimizing of power.This state makes described assembly can continue operation, but the optimum performance of described assembly and the whole abilities of any related power consumption annex may not can be provided.Second kind of described three kinds of intermediate state of operation is called hang-up.This state makes described assembly can continue operation under low power state.In order to select described suspended state, described assembly must have battery or keep communicating by letter with power supply.In order to make described assembly be back to total power from described suspended state, the operator must be to described assembly resume power.Described recuperation makes described assembly return capacity operation, and does not need to restart described assembly.The third intermediate state of operation is called dormancy.Under described dormant state, the full memory content of described computation module is written into the storer that is connected on the network, and has sent mark to described operating system: do not reset when recovering described assembly.The described memory content that described mark will indicate described operating system retrieval to store in the described storer that is connected on the network.This makes that described assembly is back to appropriateness or full power state of operation can be faster than from resetting the speed that described assembly provides.Described dormant state require described assembly before stopping power supply with its internal memory write store in.When described operator wishes described assembly when described dormant state is returned total power, they must recover described assembly.The process of recovering described assembly comprises that retrieval stores described machine disk in the past into or be connected to memory content in the storer on the network.Described three kinds of intermediate state of operation make described operator can return total power afterwards, and powered-down is fully reset described assembly afterwards again.Under every kind of described intermediate state of operation, the power that the described selected power consumption annex of described assembly and described assembly consumes is all less.So, but described three kinds of intermediate state of operation make the operator of described system can manage the power that is assigned to described subregion assembly neatly in described computing environment.
There are several embodiment can make the described computation module in the described computing system can enter one of described three kinds of intermediate state of operation.In one embodiment, can the calling system management software controlling the running status of described assembly, and the power consumption annex in the assembly.Fig. 3 is flow process Figure 100, has showed the process of Management Calculation assembly power supply on the granularity rank, the power consumption assembly of promptly described assembly and described assembly.First step of power management is to judge whether described assembly needs operation 104 continuously.This comprises the power consumption of analyzing described assembly and the operating position of judging described assembly in the described system.If do not need operating said assembly, so just require to select low-power or the no power state of operation 116 expected.Described state can comprise the power supply 118 of closing described assembly or described assembly is placed low power state 120.Operating said assembly if desired, next procedure need to judge whether any described power consumption annex 106 of operation exactly, and promptly whether they use.The power consumption annex that does not need to move can be selectively closed 108.Yet if any described assembly annex need be in running status, so just must judge whether that any described annex can operate in reduces under the capability state 110.These selected power consumption annexes can slow down 112 on basis separately then.For example, the processor that can operate in two GHz speed can be slowed down to 500 megahertzes to save energy.Can not move under the reduction capability state if judge described power consumption annex, they just must all remain under the total power and move 114 so.So the described system management software can manage the power supply of each assembly in many computation modules system, and the power supply of power consumption annex in each assembly is carried out granular management.
Described power management software can be controlled the operation of each assembly in multi-component system.As described above, each assembly in the described system can operate under the low power state, i.e. step 120.Described low power state is called dormancy and hang-up.Fig. 4 is flow process Figure 150, has showed the process of selecting the assembly low power operating state.The part of described selection running status comprises judges whether described assembly has battery 154.If described assembly does not have battery, must judge so whether described assembly will continue to communicate by letter 156 with external power source.In order to make described assembly in the time durations that prolongs, remain on suspended state, need be in the power supply during the described hang-up.For example, if described assembly will be retained in the described frame, it just can be at it continues between the suspended state duration to receive power supply from described frame.If the inquiry in step 156 is sure, described assembly just can enter described suspended state of operation 158.Close the processor power supply 160 of described assembly then, the internal memory of described assembly is maintenance activity 162 then.In described suspended state, described assembly continues power consumption, so need enough power supplys are provided or continue power supply from external power source from its internal cell.In order to make described assembly get back to running status, described assembly must pass through recuperation.This process makes described assembly can promptly return total power, and needn't restart described assembly.So, if described assembly does not remain on low power state in the time durations that prolongs, and need promptly return full power state of operation, so just may need the low power state of described hang-up.
Fig. 4 has also showed the process that assembly is placed the dormancy low power state.In this state, do not require that described assembly has battery or communicates by letter with external power source.When described assembly enters described dormant state 164, described assembly writes described machine disk with its memory content or is connected to storer 166 on the network.Described then assembly sends mark to described operating system, and the internal memory that indicates described assembly places described machine disk or is connected to storer 168 on the network.This makes described assembly visit its reflection when dormant state is finished described rejuvenation.Then to whole power consumption annex powered-downs 170 of described processor and described assembly.In a single day described assembly is in dormant state, and it just can remove and need not consider described assembly is powered continuously from described frame.Described assembly can keep dormant state at the time durations of random length.When powering for described assembly once more to make described assembly return all or part of power rating with intention, described dormancy assembly must pass through rejuvenation.Described rejuvenation is consuming time more than the described recuperation that is associated with described suspended state, but lacks than restarting of described assembly is consuming time.So, may need the low power state of described dormancy for long-term low power operating state.
If computation module in the multi-component system and the memory communication that is connected on the network, described assembly just can use disk to store as this machine alternatively.In another embodiment of described dormant state, do not comprise that the assembly of this machine hard disk also can utilize this low power state.When the assembly that does not have hard disk initially entered described dormant state, it was stored in its reflection in the described memory devices that is connected on the network by the network address in the nonvolatile memory.There is not the computation module of this machine hard disk that chip or software need be arranged on assembly, so that described operating system is simulated this machine disk.This chip or software make can communicate by letter between described assembly and the described memory devices that is connected on the network.So described chip or software make described assembly can utilize described dormant state, and need on described assembly hard disk be set.
When assembly is in or when suspended state or dormant state, described assembly can be removed from described frame.This just makes that the assembly that is in one of these two kinds of low power operating states can move to another slot or another frame, the i.e. heat interchange of same frame from a slot of described frame.In a single day described assembly places desired position in the described frame, described assembly just can or by getting back to capacity operation from the recuperation of described suspended state or by rejuvenation from described dormant state.Because these two kinds of processes are all fast than restarting fully of described assembly, when selecting described assembly to do heat exchanging process, they are more welcome.So, under the situation of component heat exchange, will return the needed time of capacity operation according to time span and the described assembly that described assembly will remain on low power state, may need described assembly is placed the low power state of described hang-up or dormancy.
Except the computation module at low power state can carry out heat interchange, can also during described restore funcitons, not need to restart described assembly and from the described different reflections of memory devices visit that are connected on the network at the computation module of dormant state.When assembly entered described dormant state, the full memory content write the described memory devices that is connected on the network from this machine disk or SCSI expansion, and sends mark to described operating system: do not restart described assembly when described component recovery is powered.In addition, described mark as indicator, indicates the position of described memory content in the described memory devices that is connected on the network to described assembly.When described component recovery was powered, described assembly can be visited the alternate image in the described memory devices that is connected on the network alternatively.As everyone knows, under the multicompartment environment, different assemblies are carried out different tasks.For example, carry out different tasks based on the Linux reflection with videoing based on Windows.Described Linux reflection and described Windows reflection can both be stored in the described memory devices that is connected on the network in dormant state.When having determined the conversion work load, can return to the described assembly that returns from described dormant state to alternate image.Alternate image can be recovered as required, comprises the computing system of subregion subregion again.The process of recovering alternate image restarts and/or reconfigures computation module faster than being higher than with efficient significantly.So, use hibernate image to shorten the switching time of recovering between some system images.
The advantage that is better than prior art
The described low power operating state of each computation module, and the granular management of the interior power consumption annex of assembly has improved the operational efficiency of multicompartment computing system.On a small scale and can managing electrical power on extensive.Intrasystem each assembly may operate in one of five kinds of ranks, is arranged as from the inactivity to the total power, has three kinds of middle operation ranks.In addition, one of described middle operation rank can be carried out granular management to stand-alone assembly.The granular management of described assembly comprises the operation of each power consumption annex of controlling described assembly.Described granular management can reduce power consumption and thermal value, can improve system configuration for replacing operating load in conjunction with described software management system during the low power applications stage, and can arrange the maintenance of each assembly in the multi-component system.
In assembly the granular management of some annexes, described power management also make or hang-up or dormant state under assembly can heat interchange.Perhaps, can place another slot in the described frame to described assembly by restoring or described assembly being returned before the total power by rejuvenation, or another frame.At last, the dormancy component recovery is during to capacity operation, moves under can be when being placed in described dormant state with it the different reflections.When recovering the assembly of dormancy, software management system can be specified the alternate image of described assembly to the described memory devices that is connected on the network.This has just improved the dirigibility of described assembly according to described system works load, and needn't restart to different reflections to described assembly fully.So power management disclosed herein has improved operational efficiency for the multicompartment computing system.
Alternate embodiment
Should admit, although for purpose this paper of showing by the agency of specific embodiment of the present invention, under the situation that does not break away from essence of the present invention and scope, can make multiple modification.Specifically, each computation module of described system can comprise hand switch, uses together with use or independence with software management system.Fig. 5 has showed multicompartment computer system 180.In this width of cloth figure, single component 184 is shown as and is placed in the frame 182.Described assembly 184 has printed circuit board (PCB) 186, comprises local battery 188 and hard disk 190.Described assembly 184 has the communication port (not shown), and described assembly can be communicated by LAN (Local Area Network) 194 and the memory devices 192 that is connected on the network.Described assembly 184 has pair of buttons 196 and 198 in its place ahead.First button 196 is used for described assembly is placed dormancy, and second button 198 is used for described assembly is placed hang-up.When described assembly operating during in total power, described button 196 and 19 will all light first kind of color, indicate flat-out running status.Described button 196,198 will light second kind of color when one of described button 196,198 person of being operated or external pressure source activation this moment, indicate described assembly and be in the process that changes its running status.In case described assembly has been finished the conversion to desired running status, described button 196,198 will light the third color.Although described button has been shown as with color and has indicated running status, described button also can use mark associated therewith to indicate running status.Equally, also can use means of communication to replace the button of installing on the described assembly 184, with the change of start assembly mode of operation such as keyboard.So protection scope of the present invention is only limited by following claims and suitable content thereof.
Claims (24)
1. the method for managing power supply in the computation module of a microprocessor system may further comprise the steps:
(a) power consumption of the described computation module of analysis;
(b) judge the power consumption level of described computation module needs according to described power consumption analysis; And
(c) running status of the described computation module of modification is with the described judgement of response to the power consumption level of described needs.
2. according to the method for claim 1, further comprise the step that the granular management of power consumption annex in the described computation module is controlled.
3. according to the method for claim 2, wherein, described power consumption annex is selected from down the group of lising composition: microprocessor, disc driver, I/O port, chip and combination thereof.
4. according to the process of claim 1 wherein, the step of revising described assembly operating state comprises the power consumption that changes described computation module by switch.
5. according to the process of claim 1 wherein, the step of revising described computation module running status comprises the described running status of controlling described computation module by software management system.
6. according to the process of claim 1 wherein, the described running status of described computation module is selected from down the group of lising composition: the selectivity minimizing of peak power, hardware accessories, hang-up, dormancy, recovery, recover, close and make up.
7. according to the method for claim 1, further comprise when described assembly when low power operating state returns, the step of visit alternate image.
8. according to the method for claim 7, wherein, the step of described visit alternate image comprises when low power state recovers described assembly, and visit is connected to the alternate image in the storer on the network.
9. according to the method for claim 1, further comprise when described computation module when low power state moves, the step of the described assembly of heat interchange.
10. according to the method for claim 9, wherein, described low power state is selected from down the group of lising composition: hang up and dormancy.
11. a multiple processor computation system comprises:
A plurality of computation modules;
Be suitable for analyzing the manager of described module dissipation, wherein, described manager is suitable for responding described analysis, calls the modification of described assembly operating state.
12. according to the system of claim 11, wherein, the modification of described running status comprises the granular management of described module dissipation annex.
13. according to the system of claim 12, wherein, described annex is selected from down the group of lising composition: microprocessor, disc driver, I/O port, chip and combination thereof.
14. according to the system of claim 11, wherein, the running status of described assembly is selected from down the group of lising composition: the selectivity minimizing of peak power, power, hang-up, dormancy, recovery, recover, close and make up.
15., further comprise storage manager, be suitable for indicating described assembly during restore funcitons, the alternate image that visit is stored in the network storage equipment according to the system of claim 11.
16. according to the system of claim 11, wherein, described assembly is suitable for when it is in low power state, carries out heat exchanging process.
17. according to the system of claim 16, wherein, described low power state is selected from down the group of lising composition:
Hang up and dormancy.
18. one kind manufactures a product, comprising:
The computer-readable signal bearing medium;
Manager is used for analyzing the power consumption of many computation modules computing environment computation module;
Control system in the described medium responds described manager, changes the running status of computation module.
19. according to manufacturing a product of claim 18, wherein, described medium is selected from down the group of lising composition: the carrier signal after recordable data storage medium and the modulation.
20. according to manufacturing a product of claim 18, wherein, the described running status of described computation module is selected from down the group of lising composition: the selectivity minimizing of peak power, hardware accessories, hang-up, dormancy, recovery, recover, close and make up.
21. according to manufacturing a product of claim 20, wherein, described control system is suitable for controlling the management of power consumption annex in the described computation module.
22. according to manufacturing a product of claim 21, wherein, described power consumption annex is selected from down the group of lising composition: microprocessor, disc driver, I/O port, chip and combination thereof.
23. according to manufacturing a product of claim 18, wherein, when described assembly when dormant state is returned, described control system is that described computation module visit is connected to the alternate image in the storer on the network.
24. according to manufacturing a product of claim 18, wherein, when described computation module just when low power state moves, described control system is suitable for providing the heat interchange of described assembly.
Applications Claiming Priority (3)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| US10/334,293 US7290152B2 (en) | 2002-12-30 | 2002-12-30 | Method and system for managing power within a compute component of a multiprocessor system |
| US10/334,293 | 2002-12-30 | ||
| PCT/GB2003/005221 WO2004059452A2 (en) | 2002-12-30 | 2003-12-01 | Power management system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1742249A true CN1742249A (en) | 2006-03-01 |
| CN1742249B CN1742249B (en) | 2010-04-14 |
Family
ID=32655012
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN200380107857.3A Expired - Fee Related CN1742249B (en) | 2002-12-30 | 2003-12-01 | Multi-processor computer system and power management method in the computing component |
Country Status (8)
| Country | Link |
|---|---|
| US (2) | US7290152B2 (en) |
| EP (1) | EP1579303A2 (en) |
| JP (1) | JP4426976B2 (en) |
| CN (1) | CN1742249B (en) |
| AU (1) | AU2003285565A1 (en) |
| CA (2) | CA2657981C (en) |
| IL (1) | IL169493A0 (en) |
| WO (1) | WO2004059452A2 (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102204226B (en) * | 2008-11-05 | 2014-10-22 | 西门子通讯公司 | Power management method and device for a communications system |
| CN104541228A (en) * | 2012-09-10 | 2015-04-22 | 英特尔公司 | Providing support for device states |
Families Citing this family (27)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6760464B2 (en) * | 2000-10-11 | 2004-07-06 | Digimarc Corporation | Halftone watermarking and related applications |
| US7290152B2 (en) * | 2002-12-30 | 2007-10-30 | International Business Machines Corporation | Method and system for managing power within a compute component of a multiprocessor system |
| US7174451B2 (en) * | 2003-03-31 | 2007-02-06 | Intel Corporation | System and method for saving and/or restoring system state information over a network |
| US7308591B2 (en) * | 2004-12-16 | 2007-12-11 | International Business Machines Corporation | Power management of multi-processor servers |
| US10474213B1 (en) * | 2005-05-30 | 2019-11-12 | Invent.Ly, Llc | Predictive power management in a wireless sensor network using scheduling data |
| US20070124573A1 (en) * | 2005-10-28 | 2007-05-31 | Walker Phillip M | Method for rapid startup of a computer system |
| US8250382B2 (en) * | 2007-08-22 | 2012-08-21 | International Business Machines Corporation | Power control of servers using advanced configuration and power interface (ACPI) states |
| US8074581B2 (en) | 2007-10-12 | 2011-12-13 | Steelcase Inc. | Conference table assembly |
| JP2009116503A (en) * | 2007-11-05 | 2009-05-28 | Buffalo Inc | Network connecting type device and program |
| US9501124B2 (en) * | 2008-05-22 | 2016-11-22 | Microsoft Technology Licensing, Llc | Virtual machine placement based on power calculations |
| US8914653B2 (en) * | 2008-09-05 | 2014-12-16 | Hewlett-Packard Development Company, L.P. | Method and system for providing hybrid-shutdown and fast startup processes |
| US20140361954A1 (en) | 2013-06-07 | 2014-12-11 | Lewis Epstein | Personal control apparatus and method for sharing information in a collaboration workspace |
| US10631632B2 (en) * | 2008-10-13 | 2020-04-28 | Steelcase Inc. | Egalitarian control apparatus and method for sharing information in a collaborative workspace |
| JP5259725B2 (en) * | 2008-10-31 | 2013-08-07 | 株式会社日立製作所 | Computer system |
| GB2468137A (en) * | 2009-02-25 | 2010-09-01 | Advanced Risc Mach Ltd | Blade server with on board battery power |
| US10884607B1 (en) | 2009-05-29 | 2021-01-05 | Steelcase Inc. | Personal control apparatus and method for sharing information in a collaborative workspace |
| JP4691180B2 (en) * | 2009-06-30 | 2011-06-01 | 株式会社東芝 | Information processing device |
| US20110145612A1 (en) * | 2009-12-16 | 2011-06-16 | International Business Machines Corporation | Method and System to Determine and Optimize Energy Consumption of Computer Systems |
| US8806237B2 (en) | 2010-03-30 | 2014-08-12 | Hewlett-Packard Development Company, L.P. | Power switch control circuitry isolating or coupling a system power bus of a computer based on a result state of a logical operation |
| US20110265090A1 (en) * | 2010-04-22 | 2011-10-27 | Moyer William C | Multiple core data processor with usage monitoring |
| US8380860B2 (en) * | 2010-11-09 | 2013-02-19 | International Business Machines Corporation | Reducing carbon footprint and providing power savings in session initiated protocol conferencing |
| CN102487542A (en) * | 2010-12-01 | 2012-06-06 | 宏碁股份有限公司 | Wireless network device with power supply management mechanism and power supply management method thereof |
| US9244521B2 (en) * | 2012-12-26 | 2016-01-26 | Intel Corporation | Supporting runtime D3 and buffer flush and fill for a peripheral component interconnect device |
| US10234919B2 (en) * | 2015-09-21 | 2019-03-19 | Microsoft Technology Licensing, Llc | Accessory-based power distribution |
| US10264213B1 (en) | 2016-12-15 | 2019-04-16 | Steelcase Inc. | Content amplification system and method |
| EP3407289A1 (en) * | 2017-05-24 | 2018-11-28 | Advanced Digital Broadcast S.A. | System and method for managing electric power consumption |
| CN107657392B (en) * | 2017-10-26 | 2021-01-08 | 燕山大学 | Particle calculation method for large-scale economic dispatching problem of power grid |
Family Cites Families (25)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5396635A (en) | 1990-06-01 | 1995-03-07 | Vadem Corporation | Power conservation apparatus having multiple power reduction levels dependent upon the activity of the computer system |
| US5452401A (en) * | 1992-03-31 | 1995-09-19 | Seiko Epson Corporation | Selective power-down for high performance CPU/system |
| JPH07200112A (en) | 1993-12-15 | 1995-08-04 | Internatl Business Mach Corp <Ibm> | Information processing system |
| US5702208A (en) * | 1994-06-02 | 1997-12-30 | Hilfiker; William K. | Grid-locked block panel system |
| US5566297A (en) | 1994-06-16 | 1996-10-15 | International Business Machines Corporation | Non-disruptive recovery from file server failure in a highly available file system for clustered computing environments |
| US5867704A (en) | 1995-02-24 | 1999-02-02 | Matsushita Electric Industrial Co., Ltd. | Multiprocessor system shaving processor based idle state detection and method of executing tasks in such a multiprocessor system |
| US5996083A (en) * | 1995-08-11 | 1999-11-30 | Hewlett-Packard Company | Microprocessor having software controllable power consumption |
| JPH09138716A (en) * | 1995-11-14 | 1997-05-27 | Toshiba Corp | Electronic computer |
| US5711691A (en) * | 1996-05-13 | 1998-01-27 | Air Packaging Technologies, Inc. | Self-closing and self-sealing valve device for use with inflatable structures |
| US5802305A (en) | 1996-05-17 | 1998-09-01 | Microsoft Corporation | System for remotely waking a sleeping computer in power down state by comparing incoming packet to the list of packets storing on network interface card |
| US6000036A (en) * | 1996-07-17 | 1999-12-07 | International Business Machines Corp. | Logical steering to avoid hot spots on integrated circuits |
| US6105089A (en) | 1997-05-13 | 2000-08-15 | Micron Electronics, Inc. | Data management system for adding or exchanging components on a running computer |
| US6134615A (en) * | 1997-05-13 | 2000-10-17 | Micron Electronics, Inc. | System for facilitating the replacement or insertion of devices in a computer system through the use of a graphical user interface |
| US6058445A (en) | 1997-05-13 | 2000-05-02 | Micron Electronics, Inc. | Data management method for adding or exchanging components on a running computer |
| JPH11161385A (en) * | 1997-11-28 | 1999-06-18 | Toshiba Corp | Computer system and its system state control method |
| US6098158A (en) * | 1997-12-18 | 2000-08-01 | International Business Machines Corporation | Software-enabled fast boot |
| US6101601A (en) * | 1998-04-20 | 2000-08-08 | International Business Machines Corporation | Method and apparatus for hibernation within a distributed data processing system |
| US6226667B1 (en) | 1998-05-26 | 2001-05-01 | International Business Machines Corporation | Method and apparatus for preloading data in a distributed data processing system |
| US6604201B1 (en) * | 1998-10-28 | 2003-08-05 | Matsushita Electric Industrial Co., Ltd. | Network unit with power saving mode inhibit based on interconnection relationship to neighboring nodes which is stored on the unit |
| EP1037133A1 (en) | 1999-03-15 | 2000-09-20 | International Business Machines Corporation | Method and apparatus for alternation between instances of operating systems in computer systems |
| US6711691B1 (en) * | 1999-05-13 | 2004-03-23 | Apple Computer, Inc. | Power management for computer systems |
| JP2001109729A (en) * | 1999-10-12 | 2001-04-20 | Nec Corp | Device and method for controlling power consumption in multiprocessor system |
| US6836849B2 (en) * | 2001-04-05 | 2004-12-28 | International Business Machines Corporation | Method and apparatus for controlling power and performance in a multiprocessing system according to customer level operational requirements |
| US6901522B2 (en) * | 2001-06-07 | 2005-05-31 | Intel Corporation | System and method for reducing power consumption in multiprocessor system |
| US7290152B2 (en) * | 2002-12-30 | 2007-10-30 | International Business Machines Corporation | Method and system for managing power within a compute component of a multiprocessor system |
-
2002
- 2002-12-30 US US10/334,293 patent/US7290152B2/en not_active Expired - Lifetime
-
2003
- 2003-12-01 AU AU2003285565A patent/AU2003285565A1/en not_active Abandoned
- 2003-12-01 CN CN200380107857.3A patent/CN1742249B/en not_active Expired - Fee Related
- 2003-12-01 CA CA2657981A patent/CA2657981C/en not_active Expired - Fee Related
- 2003-12-01 JP JP2004563328A patent/JP4426976B2/en not_active Expired - Fee Related
- 2003-12-01 CA CA2509943A patent/CA2509943C/en not_active Expired - Fee Related
- 2003-12-01 EP EP03778563A patent/EP1579303A2/en not_active Withdrawn
- 2003-12-01 WO PCT/GB2003/005221 patent/WO2004059452A2/en active Application Filing
-
2005
- 2005-06-30 IL IL169493A patent/IL169493A0/en unknown
-
2007
- 2007-07-20 US US11/781,087 patent/US7596686B2/en not_active Expired - Fee Related
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102204226B (en) * | 2008-11-05 | 2014-10-22 | 西门子通讯公司 | Power management method and device for a communications system |
| CN104541228A (en) * | 2012-09-10 | 2015-04-22 | 英特尔公司 | Providing support for device states |
| US9720701B2 (en) | 2012-09-10 | 2017-08-01 | Intel Corporation | Providing support for device states |
Also Published As
| Publication number | Publication date |
|---|---|
| CA2657981C (en) | 2011-09-06 |
| WO2004059452A2 (en) | 2004-07-15 |
| CA2657981A1 (en) | 2004-07-15 |
| CA2509943C (en) | 2010-07-27 |
| US7596686B2 (en) | 2009-09-29 |
| AU2003285565A8 (en) | 2004-07-22 |
| US20040128564A1 (en) | 2004-07-01 |
| CN1742249B (en) | 2010-04-14 |
| US20070266266A1 (en) | 2007-11-15 |
| AU2003285565A1 (en) | 2004-07-22 |
| EP1579303A2 (en) | 2005-09-28 |
| US7290152B2 (en) | 2007-10-30 |
| JP4426976B2 (en) | 2010-03-03 |
| CA2509943A1 (en) | 2004-07-15 |
| JP2006512650A (en) | 2006-04-13 |
| IL169493A0 (en) | 2007-07-04 |
| WO2004059452A3 (en) | 2004-10-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN1742249B (en) | Multi-processor computer system and power management method in the computing component | |
| Barroso et al. | The datacenter as a computer: An introduction to the design of warehouse-scale machines | |
| CN100458657C (en) | Method and system of power management of multi-processor servers | |
| CN102326132B (en) | Power management to maximize reduced power state for virtual machine platforms | |
| US7228441B2 (en) | Multi-server and multi-CPU power management system and method | |
| US8700923B2 (en) | Apparatus and method for modular dynamically power managed power supply and cooling system for computer systems, server applications, and other electronic devices | |
| CN1308793C (en) | Method and system for machine memory power and availability management | |
| US7237129B2 (en) | System and method for activity or event based dynamic energy conserving server reconfiguration | |
| US7051215B2 (en) | Power management for clustered computing platforms | |
| US7877358B2 (en) | Replacing system hardware | |
| US7058826B2 (en) | System, architecture, and method for logical server and other network devices in a dynamically configurable multi-server network environment | |
| US8145924B2 (en) | Storage apparatus and start-up control method for the same | |
| US20120272243A1 (en) | Protecting high priority workloads in a virtualized datacenter | |
| US20070245165A1 (en) | System and method for activity or event based dynamic energy conserving server reconfiguration | |
| CN101470518A (en) | Method and device for service independent of os | |
| Barroso et al. | The datacenter as a computer | |
| US9141172B1 (en) | Method and apparatus to manage and control a power state of a device set based on availability requirements of corresponding logical addresses | |
| EP4027241A1 (en) | Method and system for optimizing rack server resources | |
| US8122268B2 (en) | Reducing power consumption of mirrored RAID subsystems | |
| Dar et al. | Power management and green computing: an operating system prospective | |
| Intro et al. | The Datacenter as a Computer |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| C41 | Transfer of patent application or patent right or utility model | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20161216 Address after: Singapore Singapore Patentee after: Lenovo Enterprise Solutions (Singapore) Pte. Ltd. Address before: American New York Patentee before: International Business Machines Corp. |
|
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20100414 Termination date: 20171201 |